GLYCOARRAY SECTION

Multi-dimensional glycan microarrays with glyco-macroligands

Satya Nandana Narla1 & Huan Nie2 & Yu Li2 & Xue-Long Sun1

Received: 28 January 2015 /Revised: 6 March 2015 /Accepted: 10 March 2015 /Published online: 10 May 2015# Springer Science+Business Media New York 2015

Abstract Glycan microarray has become a powerful high-throughput tool for examining binding interactions of carbo-hydrates with the carbohydrate binding biomolecules like pro-teins, enzymes, antibodies etc. It has shown great potential forbiomedical research and applications, such as antibody detec-tion and profiling, vaccine development, biomarker discovery,and drug screening. Most glycan microarrays were made withmonovalent glycans immobilized directly onto the array sur-face via either covalent or non-covalent bond, which afford amultivalent glycans in two dimensional (2D) displaying. Avariety of glyco-macroligands have been developed to mimicmultivalent carbohydrate-protein interactions for studyingcarbohydrate-protein interactions and biomedical researchand applications. Recently, a number of glyco-macroligandshave been explored for glycan microarray fabrication, in par-ticular to mimick the three dimensional (3D) multivalent dis-play of cell surface carbohydrates. This review highlights the-se recent developments of glyco-macroligand-based microar-rays, predominantly, novel glycan microarrays with glyco-macroligands like glycodendrimers, glycopolymers,glycoliposomes, neoglycoproteins, and glyconanoparticleswith the effort in controlling the density and orientation ofglycans on the array surface, which facilitate both their bind-ing specificity and affinity and thus the high performance ofglycan microarrays.

Keywords Glycanmicroarray . Glycodendrimers .

Glycopolymers . Glycoliposomes . Neoglycoproteins .

Glyconanoparticles

Introduction

Cell surface carbohydrates, prevailing as glycoconjugatessuch as glycoproteins, glycolipids and proteoglycans, are in-volved in many biological processes, such as cell-cell interac-tion [1, 2] immune recognition events [3], pathogen-host in-teraction [4], tumor metastasis [5], tissue growth and repair [6]etc.. Understanding the interactions between carbohydratesand carbohydrate binding proteins (CBPs) are highly neededfor clarifying the molecular mechanisms of these physiologi-cal and pathological pathways and discovering new therapeu-tic and diagnostic principles. Glycan microarrays have beendeveloped and proved as powerful high throughput tools tostudy the interactions between carbohydrates and carbohy-drate binding molecules like proteins [7, 8], antibodies [9],and pathogens [10]. Recently, they have been applied to clin-ical antibody detection and profiling [11], biomarker discov-ery [12], and drug screening [13] applications. Nevertheless,many challenges still exist for their potential biomedical re-search and applications. Mainly, there are two criticallimitations that prevent the wide and practical applications ofglycan microarray technology. First, the detection is limitedby limited epitope availability for microarray fabrication fromboth isolation from natural sources and synthesis. Both isola-tion and synthesis of carbohydrates are still very challengingdue to their structure complexity, low throughput, intense la-bor as well as the need of special skills. Second, the features ofglycan presentation on the microarray surface such as densityand orientation of glycans have substantial effect on proteinrecognition related to both affinity and specificity.

* Xue-Long Sunx.sun55@csuohio.edu

1 Department of Chemistry, Chemical and Biomedical Engineeringand Center for Gene Regulation in Health and Disease (GRHD),Cleveland State University, 2121 Euclid Avenue,Cleveland, OH 44115, USA

2 Department of Life Science and Engineering, Harbin Institute ofTechnology, Harbin, China 150001

Glycoconj J (2015) 32:483–495DOI 10.1007/s10719-015-9580-z

Conventional glycan microarrays were developed as twodimensional (2D) glycan ligands with monovalent glycansimmobilized directly onto the array surface via either covalentor non-covalent bond [7–10], which affords the multivalentglycans but lacks of the three dimensional (3D) cell surfacecarbohydrates formats. In addition, there is little control overglycan presentation. Therefore, these glycan arrays usuallyresult in low signal intensity and substantial non-specific bind-ing of target proteins due to an insufficient number of acces-sible glycans and the presence of surface-protein interactions.To overcome these limitations, recently, glycan density andorientation on the microarray surface has been recognized asan important feature of carbohydrate recognition and thus hasbeen investigated extensively. A recent comprehensive reviewpaper summarized well the most important development ofglycan array technologies [14]. Overall, different glycan arrayplatforms produced different results, which is well demon-strated by a comparative analysis of glycan array platformswith six different modes of glycan presentation using fivewell-known lectins reported by Mahal et al. [15]. As theyindicated, multiple microarray platforms may be required toobtain the best understanding of glycan binding in a systemand much microarray development remains to get a compre-hensive view of glycan-binding protein specificity.

In nature, carbohydrate receptors exist mostly asglycoconjugates such as glycoproteins and glycolipids, whichpresent multivalent carbohydrates either in a singleglycoconjugate or by combining multiple glycoconjugates asa cluster. In addition, CBPs contain two or more carbohydratebinding sites or assemble into oligomers withmultiple bindingsites, which allow them to simultaneously bind two or morecarbohydrates and facilitate functional affinity (avidity) andselectivity. Therefore, multivalent carbohydrate ligand presen-tation has been a central theme to study and apply intricate

carbohydrate-protein interactions. In the past decades, variousglycopolymeric molecules have been investigated for study-ing carbohydrate-protein interactions and biomedical researchand applications, such as for multivalent inhibitors and antag-onists development. For example, different multivalent carbo-hydrate molecules like glycodendrimers [16–22],glycopolymers [23–28], glycoliposomes [29–32], neo-glycoproteins [11, 13, 33], and glyconanoparticles [34] asglyco-macroligands were developed to attain more specificand strong carbohydrate-protein interactions. It is acknowl-edged that not only the multivalency of the carbohydratesaffect the binding specificity but also the orientation of thecarbohydrates on the solid surface affects the interactions sig-nificantly, i.e. the tight binding of proteins to carbohydrates onthe array surface depends on multivalency of carbohydrateswith appropriate spacing and orientation of carbohydrates onthe surface. Recently, many glyco-macroligands have beenused for glycan microarray fabrications, in particular with ef-fort to control the density and orientation of the glycans on themicroarray surface to thoroughly represent multi-dimensionalcell surface carbohydrates closely. These novel approaches tocontrol multivalency, density and the orientation of carbohy-drates on the microarray surfaces have shown great potentialfor biomedical research and applications. This review high-lights these recent developments of glyco-macroligand-basedmicroarrays, primarily focusing on the fabrication of novelmulti-dimensional glycan arrays with glyco-macroligandssuch as glycodendrimers, glycopolymers, glycoliposomes,neoglycoproteins, and glyconanoparticles and their potentialbiomedical applications as well (Fig. 1). In particular, how themultivalency, density and orientation of these glycans werecontrolled on the microarray surface and their impact onligand binding to the carbohydrates presentation arediscussed.

Fig. 1 Multi-dimensional glycan arrays with glyco-macroligands such as glycodendrimers, neo-glycoproteins, glycopolymers, glycoliposomes, andglyconanoparticles

484 Glycoconj J (2015) 32:483–495

Glycodendrimer-based glycan microarray

Dendrimers are a class of highly branched, monodisperseglobular macromolecules [16–18, 35, 36], which are charac-terized as highly branched ‘dendrons’ or ‘wedges’ that origi-nate from one central multifunctional core unit and terminatein many functional groups around their peripheries [16]. Sev-eral research groups have confirmed the potential to usedendrimers as scaffolds for multivalent carbohydrate ligandsby synthesizing carbohydrate containing dendrimers oftentermed as glycodendrimers [17, 18]. Recently, thedistinguishingmultivalency of these glycodendrimers attainedattention in synthesis of glycodendrimers for microarray ap-plication [19–22].

Pieters and coworkers reported glycodendrimer-based gly-can microarray and studied the binding patterns of lectins tocarbohydrates with respect to the multivalency of carbohy-drate moieties and the inter-binding-site distances of lectins[19, 20]. They designed mannose-based glycodendrimers andtested against two different lectins Con A and WGA [19],later, extended their work with five different glycodendrimersto evaluate a series of lectins [20]. Briefly, they synthesizedalkyne-functionalized dendrimers with an amine functionalgroup linked to the core which helps in immobilization ofthe dendrimers on solid surfaces. Azide functionalized sugars(Man, Glc, GlcNAc, Gal, and Galα1,4Gal), were introducedat the periphery by copper-catalyzed click chemistry followedby deprotection of sugars to afford glycodendrimers rangingfrom monovalent to octavalent dendrimers. Theseglycodendrimers were spotted on malemide-functionalizedporous aluminum oxide surface (Fig. 2A) and binding affini-ties of several FITC-lectins (ConA, GNA,WGA,GS-1, LCA,PSA, DSA, PNA, CTB5, and LecA) were studied by flow-through microarray technology. They tested the binding affin-ity of lectins Con A and GNA to mannose-basedglycodendrimers microarray with varying concentrationsranging from 0.1 to 5 mM and valences of mono, di-, tetra-and octa-dendrimers. It was observed that the lectin Con Ashowed no major multivalency effect supporting the fact that,Con A is a tetramer with binding sites that are separated byover 60 Å and are too far apart to binding over dendrimers.

However, lectin GNA with 12 binding sites spaced with adistance of 20 Å showed significant increase in the bindingto mannose dendrimers with increase in the valence ofdendrimers from mono to octa. In their further studies [20],they investigated the multivalency effect on both binding af-finity and specificity towards lectins. Lectin WGA with twoclosely spaced binding sites showed strong multivalent effect;also the binding was observed only to GlcNAc-glycodendrimer at high valence and not to otherglycodendrimers. The same phenomenon was observed withthe other eight lectins that were tested against five differentglycodendrimers, lectins with increase in inter–binding-sitedistance displayed decrease in the multivalency effect.

Wang and coworkers employed an altered approach forconstruction of glycodendrimer-based microarray by initiallypreparing a three dimensional dendrimeric platform followedby attaching the carbohydrate to the dendrimer surface [21].Briefly, the poly(amido amine) (PAMAM) dendrimersimmobilized glass slides were functionalized with hydrazideand aminoxy functional groups, followed by microspottingthe carbohydrates dissolved in different buffers under micro-wave radiation energy (Fig. 2B). The PAMAM dendrimerstested in this study included five generations G1, G2, G3,G4 and G5 containing 8, 16, 32, 64 and 128 surface groups,respectively. For the detection of carbohydrate-lectin interac-tions, direct immunoassay was performed with individual ormixed Cy3-labeled lectins and sandwich immunoassay wasperformed with individual or biotinylated lectins and Cy3-streptavidin. The cluster effect was tested by spotting oligo-saccharides GlcNAcβ1-4GlcNAc, GalNAc and GlcNAc ontoaminoxy and hydrazide functionalized PAMAM dendrimersof all five generation, which were further incubated with Cy3-labeled WGA. As a result, with the increase in the surfacegroups the fluorescent intensity increased, indicating themultivalency effect. Also, hydrazide-dendrimer coated slidesprovided higher fluorescent signal/noise ratio compared toaminoxy-dendrimers coated slides. Further, the specificity ofdifferent glyco-epitopes was examined through a panel ofmono-, oligo- and polysaccharides by micro spotting themonto hydrazide dendrimers surface. These carbohydrate mi-croarrays were incubated with either a mixture of biotin-

Fig. 2 Glycodendrimer-based glycan arrays: A Immobilized glycodendrimers via click chemistry [20] and B immobilization of oligosaccharides ontoaminoxy-dendrimer and hydrazide-dendrimer coated glass slides [21]. Modified from Ref. [20, 21]

Glycoconj J (2015) 32:483–495 485

labeled lectins and anti-sialyl lewis x antibody or with indi-vidual lectin/antibody separately and finally with Cy3-streptavidin. The binding profiles verified the specificity ofthe carbohydrate epitopes on the hydrazide-dendrimersurfaces.

Miura and co-workers studied glycodendrimer microarraywith surface Plasmon resonance (SPR) (20) and found that thebinding kinetic constants were unique to each lectin, implyingspecificity of the multivalent effect of glycan–lectin interac-tion. The glycodendrimers with various generations and sac-charides (α-Man, β-GlcNAc and β-Gal) were synthesizedand immobilized on an acetylenyl-terminated gold substratevia click chemistry. The binding abilities of the saccharide-immobilized substrates were evaluated by SPR with thelectins of Con A, WGA and RCA120. Lectin binding to theglycodendrimer array was quantitatively analyzed, and asso-ciation constants (KA) and kinetic rate constants (ka and kd)were evaluated.

Overall, glycodendrimers were successfully applied for mi-croarray fabrication and were qualified in the screening ofcarbohydrate-protein interaction applications. Theseglycodendrimer microarrays were used not only to explorethe protein-carbohydrate interactions but also assist in under-standing the influence of multivalent carbohydrate display onthe microarray surface in improving the sensitivity and selec-t ivi ty of glycan microarrays. From the reportedglycodendrimers microarrays it is recognized that themultivalency of the carbohydrates is well controllable bydendrimers. However, the spacing and orientation of the car-bohydrates on the array surface are not yet structured bydendrimers to study their contribution to sensitivity and selec-tivity of glycan microarrays in capturing glycan binding mol-ecules . This is a major drawback of the currentglycodendrimers microarray techniques and continued studieson glycodendrimer design and its immobilization chemistryare further needed.

Neo-glycoprotein-based glycan microarray

Neo-glycoproteins are a group of glycomacroligands synthe-sized by covalently attaching glycans to proteins and showgreat potential as glycoprotein mimetics for biomedical re-search and applications [36, 37]. Neo-glycoproteins have alsobeen applied for glycan microarray fabrication and haveshown great potential for biomedical research and applica-tions. Gildersleeve et al. developed neo-glycoproteins by co-valently attaching carbohydrate moieties to lysine residues ofbovine serum albumin (BSA) for microarray applications(Fig. 3) [11–13]. The carrier protein BSA served as both themultivalent scaffold and the linker for immobilization on toglass slides. Carbohydrates were conjugated to free aminegroups of BSA via reductive amination of oligosaccharides

or by coupling the oligosaccharides with carboxylic acid viaactivation with EDC/NHS. To test the glycan density effect,they synthesized 45 BSA-glycoconjugates by varying carbo-hydrate density of 11 different glycans on BSA [11]. Theseneo-glycoproteins were immobilized onto epoxy functional-ized glass slides to test their binding patterns of plant lectinsV. villosa B4 (VVL-B4), H. Pomatia aggutinin (HBA), andsoybean agglutinin (SBA) (Fig. 2A). As a result, they ob-served that the selectivity of lectins to a particular carbohy-drate over other carbohydrates declined at high density andalso the selectivity of conjugate for an individual lectin overother lectins declined at high density. Further, they also eval-uated the density-dependent binding abilities of a set of mono-clonal antibodies and determined that antibodies to the samecarbohydrate antigen can recognize density in significantlydifferent ways. Also, they investigated the density-dependentbinding properties of serum antibodies from 30 subjects, theirresults demonstrated that modulation of glycan antigen densi-ty could be used to reveal differences in antibody populationsbetween different subjects. The results from the above studyrevealed that both structure and the density of the carbohy-drates significantly affect the affinity and selectivity to carbo-hydrate binding molecules.

Followed by this work, Gildersleeve group further devel-oped an approach to modulate the neoglycoprotein density onthe array surface [13]. Briefly, the glycan microarrays wereconstructed with approximately 600 combinations of glycansand presentat ions, with varia t ions in densi ty ofneoglycoprotein, structure of glycans and the density of gly-cans. The microarray was designed to distinguish between thecarbohydrate binding proteins that can form one-to-one com-plex (Fig. 3Ba), i.e. form a multivalent complex with a singleneoglycoprotein or one protein binding to two or moreneoglycoproteins, i.e. recognize carbohydrates on adjacentneoglycoproteins to form a bridge complex (Fig. 3Bb) andto study the effect of these bridging complexes with theneoglycoproteins spaced further apart on the surface(Fig. 2Bc). Their strategy to achieve this kind of array is toadd unmodified protein (BSA) to the solution ofneoglycoprotein and print these mixtures on the array surface,in which the unmodified BSA occupies surface between theneoglycoproteins based on the amount of unmodified BSAadded to the mixture (Fig. 3Bc). They prepared the arraysurfaces with four different ratios of BSA per array compo-nent: 1:0 (100 % BSA), 1:1 (50 % BSA), 1:3 (25 % BSA)and 1:7 (12.5 % BSA), with 147 different neoglycoproteinsand glycoproteins along with several controls to a total of 591combinations. Plant lectins concanacalin A (Con A),Viciavillosaiso lectin B4 (VVL-B4) and Riciniscommunisagglutinin (RCA120) were tested over the microarrays andobserved that the spacing between the neoglycoproteinsshowed significant influence on the binding affinity oflectins.

486 Glycoconj J (2015) 32:483–495

Later on, the same group studied the effects ofneoglycoprotein density on antibody binding [33]. Briefly,they investigated the binding affinities of five commerciallyavailable antibodies 81FR2.2, HE-195, HE-193, B480, andZ2A to several antigens and found that neoglycoprotein den-sity had a considerable effect on antibody recognition(Fig. 3C). Further, the binding affinities of antibodies fromhuman serum from 15 healthy subjects were analyzed anddetermined that the binding affinities of different subpopula-tion were different based on the density of neoglycoprotein. Inaddition, they evaluated the immune response induced byprostate cancer vaccine from six individuals before and aftervaccination and verified that the variations in neoglycoproteindensity could be used to detect antibody responses. Overall,these researches demonstrated that the density of the

carbohydrates on the array surface can be adjusted by usingneoglycoproteins with different carbohydrate density on it,which are significant in studying the carbohydrate bindingligands. These studies set up a general protocol glycan micro-array techniques using any kind of carbohydrates and screen-ing their functions and potential biomedical applications.

Glycopolymer-based glycan microarray

Glycopolymers carrying pendant sugar moieties on the poly-mer backbone as multivalent carbohydrate derivatives havebeen proven as an effective tool to study carbohydrate-basedbiological processes and have shown great potential in bio-medical research and applications [38]. Glycopolymers are

Fig. 3 Neo-glycoprotein-basedglycoarray: A neoglycoproteinwith variations in carbohydratesand density on the array surface[11], B multivalent bindingmodes and the array strategy ofneoglycoprotein [13], Cillustration of high and lowdensity neoglycoprotein on thesurface [33]. Modified from Ref.[11, 13, 33]

Glycoconj J (2015) 32:483–495 487

Fig. 4 Glycopolymer-based glycoarray: a duel end functionalized mucinlike glycopolymer with alkyne on one end and Texas Red dye on the otherend [24], b mucin like glycopolymer presentation that mimics nativemucin [40], c biotin end functionalized glycopolymer immobilizedglass slide [26], d glycopolymer formation at different illumination

times [41], e O-cyanate end functionalized glycopolymer immobilizedon amine functionalized glass slide [27], f O-cyanate chain-endfunctionalized glycopolymer pre-complexed and immobilized withboronic acid ligands of different sizes [28]. Modified from Ref. [24,26–28, 40, 41]

488 Glycoconj J (2015) 32:483–495

either synthesized by direct polymerization of carbohydrate-containing monomers with or with out protection group or bythe post polymerization glyco-conjugation of synthetic poly-mers [23]. It is generally accepted that synthetic glycopolymerscan mimic the functions of naturally occurring polysaccharides[28, 29, 38] and have been employed as binding cell-surfacereceptors [30, 38]. The strength and selectivity of binding in-teractions between multivalently displayed carbohydrates andtargets are likely to depend on the density and relative spatialarrangement of the carbohydrate residues of glycopolymers. Inrecent years, chain-end functionalized glycopolymers weresynthesized and demonstrated homogeneous immobilizationof the glycopolymer via a single terminal anchor to yield anoriented and density controlled exhibition of multivalent pen-dant carbohydrates and were used for glycan microarray appli-cations [38].

Bertozzi’s group synthesized a duel end functionalizedmucin like glycopolymer [24] with a center mucin mi-metic domain with N-acetylgalactosamine moieties, oneterminal alkyne group as surface attachment element andthe other terminal outfitted for conjugation of Texas Redas a fluorophore. By microcontact printing the alkyneterminal glycopolymers were immobilized onto azidefunctionalized surface in presence of a copper catalyst,which forms a stable triazol linkage to covalently immo-bilize the glycopolymer. As the glycopolymer was alsofunctionalized with fluorescent tag Texas Red on theother end the immobilization of polymer was confirmedby fluorescent microscopy. Also, the ligand specificity ofglycans to the polymer was tested by immobilizing non-fluorescent tag glycopolymer with Texas Red taggedHelixpomatia agglutinin (Texas Red-HPA) that recog-nizes α-GalNAc. Examination of the surface by fluores-cent microscopy revealed the specific binding of HPAonly to α-GalNAc (Fig. 4a) confirming the specificityof glycans. They later developed an approach for thesynthesis of glycopolymers for microarray applicationsvia ligation of reducing sugars to the polymer backbonescarrying hydrazide groups and end functional biotingroup [25]. They arrayed a cluster of biotinylatedglycopolymers onto streptavidin-coated glass slides,which were recognized by lectins specifically dependingon their pendant glycans.

Bertozzi et al. also synthesized duel end functionalizedmucin like glycopolymers with α-N-acetylgalactosamine (α-GalNAc) moieties via RAFT polymerization for studying thecross-linking of mucin like glycoconjugates by lectins [39].The polymer was terminated with biotin on one end to facil-itate anchoring of glycopolymer on to streptavidin-coated mi-croarray substrates and a trithiocarbonate moiety on the otherend that provided a free sulfhydryl group for conjugation of amaleimide-functionalized Cy3 dye upon rapid aminolysiswith cysteamine in DMF. They synthesized five mucin-like

glycopolymers with a range of GalNAc valencies (68, 92, 111,146, and 170). To control the density of the glycopolymer onthe surface they spotted glycopolymer solutions of differentconcentrations (400, 150, and 75 nM) on the streptavidin coat-ed glass slides. Apparent dissociation constants of four lectins(Soybean agglutinin (SBA),Wisteria floribunda lectin (WFL),Vic ia v i l losa -B-4 agglu t in in (VVA), and Hel ixpomatiaagglutin (HPA)) with specificity to α-GalNAc weredetermined. They observed valency dependent binding forSBA, WFL and VVA lectins, whereas HPA showed strongavidities toward all the polymers irrespective of GalNAcvalency. Though all four lectins have capacity to cross-linklow valency glycoconjugates, only the high valency GalNAcmucin mimetic polymer showed cross-linking to SBA, whichreveals that glycan valency and organization are critical pa-rameters to determine the cross-linking properties of lectins(Fig. 4b).

Glycosaminoglycans play important roles on cell sur-face. Hsieh-Wilson et al. synthesized a biotin end func-tionalized glycopolymer that mimic chondroitin sulfateproteoglycans via ring-opening metathesis polymerizationtechnique [26], which were employed for microarray andSPR applications. Both sulfated CS-E and non-sulfatedCS-C glycopolymers were immobilized onto streptavidincoated glass slides by high-precision contact-printing ro-bot, which delivers nanoliter volumes of glycopolymeronto glass slides. Polymer immobilized glass slides(Fig. 4c) were tested for binding of specific monoclonalantibodies 2D11 and 2D5, which are selective for CS-Eand CS-C glycopolymers, respectively. They also studiedthe binding of a growth factor, glial cell-derived neuro-tropic factors selective for sulfated epitopes, whichshowed higher selectivity for CS-E compared to CS-Cby both microarray and SPR techniques.

Most recently, Braunschweig et al. employed a new ap-proach by combining polymer pen lithiography (PPL) andbeam pen lithiography (BPL) with acrylate or methacrylatephoto polymerization chemistry to generate arrays of brushpolymers with side chain functionalized with fluorophoresor glycans [40]. They patterned acrylate and methacrylatemonomers (α-mannose and α-glucoside) onto thiol coatedglass slides and exposed these glass slides to UV light thatproduced brush polymers by a photoinduced radical acrylatepolymerization reaction. Both the monolayers and the brushpolymers were treated with fluorophore labeled Con A at dif-ferent concentrations, to investigate the architecture effect ofthese glycans in binding to Con A. The fluorescence intensityfor the brush polymers binding was observed to be 20 timesgreater than that for glycan monolayers at high concentrationsof Con A. The different polymer brush heights were attainedby variation in irradiation times (2, 5, 10 and 20 min), with126 nm height polymers at 20 min whereas 8 nm height poly-mers at 2 min, which was determined by AFM. Also, they

Glycoconj J (2015) 32:483–495 489

found that the polymers with longer length (126 nm) are anorder of magnitude more sensitive than shorter polymers(8 nm) towards binding Con A (Fig. 4d).

We recently developed an approach for glycan microar-rays using glycopolymer to control orientation and densitywith O-cyanate chain-end functionalized glycopolymers[27, 28]. Initially, we synthesized an O-cyanate chain endfunctionalized glycopolymer with pendant lactose moietieson the backbone via cynaoxyl-mediated free-radical poly-merization technique [27]. In our study, glycopolymer mi-croarray (spot size 500 μm diameter) was fabricated bymicro contact stamping glycopolymer onto amine func-tionalized glass slide and incubated the glass slides withlectin-FITC (Arachishypogae, FITC-Labeled, Sigma), thensubjected them to fluorescent imaging. A control was pre-pared by incubating the glycopolymer array with lactosepre-incubated lectin-FITC instead of free lectin. Fluores-cence microscopy analysis of the surfaces revealed specificbinding of lectin to the immobilized glycopolymer(Fig. 4e). The lectin binding to arrayed glycopolymer wasinhibited in the presence of free lactose, further confirmingthe specific lectin binding to the glycopolymer. Moreover,glycopolymer array with different concentration ofglycopolymer showed concentration-dependent lectinbinding to glycopolymer.

The glycopoymer-based microarrays showed great po-tentials as multi-dimensional glycan microarrays. Howev-er, the density of the immobilized glycopolymer on thearray surface is still uncontrollable, and thus, the possibil-ity to access the multivalent glycans in parallel maybe lim-ited and therefore does not facilitate maximum proteinbinding affinity and specificity. We proposed a novel ap-proach to control the density of glycopolymers on the arraysurface to solve this problem [28]. First, we examined theg lycan dens i ty ef fec t fo r lec t in b ind ing in theglycopolymer by synthesizing a series of glycopolymerswith different densities of pendant glycan and molecularweights. Four kinds of glycopolymer were synthesized andglycopolymer microarrays were fabricated by microcontactstamping of glycopolymers onto amine functionalizedglass slides followed by incubating with PNA lectin-FITC.Glycopolymer with 1 to 30 ratio of lactose and acryl amideshowed the highest level of lectin binding compared to theglycopolymers with 1 to 18, 1 to 54 and 1 to 51 ratio oflactose and acryl amide. These results indicated that theglycan density in the polymer has impact for the lectinbinding. Next, for controlling the density of immobilizedglycopolymer on array surface; we developed end-pointimmobilization of glycopolymer combined with molecularspacing technique. Briefly, O-cyanate chain-end function-alized glycopolymer was pre-complexed with boronic acidligands in different sizes and then immobilized onto amine-functionalized glass slides. After the immobilization, the

spacer boronic acid ligands were released from theimmobilized glycopolymers at reduced pH condition toafford the oriented and density controlled glycopolymermicroarray (Fig. 4f). We synthesized three macro-boronicacid moieties lysozyme-BA (MW: about 15 kDa), BSA-BA (MW: about 70 kDa) and polyacrylamide-BA (MW:about 10 kDa) as a spacing molecule to vary immobilizedglycopolymer density. These glass slides were then incu-bated with two different lectins PNA and RCAI, both havespecificity to β-galactose. Interestingly, the binding ofboth lectins was enhanced when the spacers were usedfor immobilization. PNA and RCAI showed higher bind-ing for different ratios of glycopolymer to BA-spacer, stat-ing that lectins that bind same glycans are affected inunique way. Further, this density-dependent lectin bindingspecificity was confirmed by SPR technique. Overall, the-se results indicated that oriented glycopolymers with con-trolled ligand density facilitates enhanced protein binding,which is very important for studying glycan-protein inter-action, such as assessing multivalent glycans in parallel ina microchip format and other solid phase assays such asbiosensors.

Glyco-membrane mimetics-based glycan microarray

Cell membrane mimetic systems play fundamental roles inunderstanding molecular interactions on the cell surface andprovide enormous opportunities in developing products forbiomedical research and applications [41]. Membrane mimet-ic systems can be designed to have controlled environmentthat mimic the characteristics of a native membrane, such ascomposition, structure, curvature and charge close to theirnative structure as in the cell membrane [42]. A variety ofmembrane mimetic systems have been developed for investi-gation of various membrane-related processes. Liposomesare one of the most common membrane mimetic sys-tems, which are small artificial vesicles of sphericalshape that can be created from cholesterol and naturalnon-toxic phospholipids [43, 44]. Lipid composition, sur-face charge, size, and the method of preparation are keyfactors that affect the liposome properties [45]. In addi-tion, supported lipid membrane system emerged as themost attractive platform because of its quasi-natural set-ting and that its size, geometry and composition can betailored with great precision to the complexity of biolog-ical membrane. Several platforms such as solid-supported[46–50], tethered [51–53], polymer cushioned [54–56]lipid bilayers and supported vesicular systems [57, 58]have been developed. So far, glyco-membrane mimeticsystems such as supported glyco-lipid membrane systemsand glyco-liposomes have been used for glycan microar-ray applications [29, 30, 59].

490 Glycoconj J (2015) 32:483–495

Supported glyco-lipid bilayer-based glycan microarray

Guo et al. demonstrated a fluidic glycan microarray for quan-titative glycomics application [29] based on robust supportedlipid bilayer (SLB) technology. This approach is to incorpo-rate tethered cholesteryl groups onto the bottom leaflet of thesupported bilayer lipid membrane to increase the stability andrigidity of the membrane, also combined with tether achievesdesired robustness (Fig. 5). Briefly, a supported lipid bilayerwas formed by spotting small unilamellar vesicle (SUV) ontothe tethered cholesterol and verified the fluidity of SLB byfluorescence recovery after photo bleaching. Different con-centrations of mannose linked lipids were incorporated inthe lipid mixture to fabricate density gradient microarray withknown mannose density of each spot. The mannose-basedmicroarray was then incubated with a suspension of E. coli.ORN178 and ORN208 as a negative control. It was found thathigh E. coli. adhesion was observed with the increase of man-nose density. The number of cells adhered to the array slidewas counted, the fluorescent microscopy of E. coli. adhesionto the SLB sports containing 0.1, 1,5 and 10 % mannose andthere is a rapid increase in the cell count at ≥5 % and reachedsaturation later on. Also, an inhibition study was conductedwhere the mannose nanoparticles were added to the suspen-sion of E. coli.As a competitive inhibitor, and with increase inthe concentration of nanoparticles the cell adhesion to themannose array decreased. This study indicated that the dy-namic clustering of mannosyl groups on the fluidic membranesurface could simulate the function of complex oligo- andhigh-mannose molecules.

Glyco-liposome-based glycan microarray

Liposome microarrays have been investigated for several ap-plications in membrane biophysics, biotechnology, and col-loid and interface science [57, 58]. We recently investigatedthe application of azide-reactive liposome for efficient andchemically selective immobilization and microarray fabrica-tion (Fig. 6a) [30]. Azide reactive liposomes were synthesizedwith DSPE-PEG2000-triphenylphosphine, phospholipids(DSPE) and cholesterol rapid extrusion method. Azide-PEG6-Biotin was conjugated to prepared liposomes in 7.4pH PBS buffer via Staudinger ligation and immobilizationof these biotinylated liposomes was performed by incubatingthem onto streptavidin functionalized glass slides. To confirmthe immobilization of intact liposomes on glass slides, DSPErhodamine dye was doped into the lipid bilayer into both postbiotinylated and directly biotinylated liposomes and the 5,6-carboxyfluorescene dye was encapsulated inside the lipo-somes and subjected to fluorescence imaging. Both rhoda-mine and the 5,6-CF tags were observed at respective wave-lengths confirming that the arrayed intact liposomes wereachieved through specific biotin/streptavidin interaction.Next, the azide reactive liposomes were immobilized on azidefunctionalized glass slides and later subjected to glycosylationby conjugation of 2-azideethyl-lactose to leftover TP on theliposome’s exterior surface. Lectin binding studies were con-ducted by incubating the liposome immobilized glass slidewith lectin PNA. The specific lectin binding to lactosylatedliposome was observed, while no lectin binding was observedto liposome with just anchor group.

Fig. 5 Formation of a glycanpresenting supported lipid bilayer(SLB) surface from a smallunilamellar vesicle (SUV)solution and glycan densitygradient microarray for pathogenadhesion [29]

Glycoconj J (2015) 32:483–495 491

Further, this work was extended to study the glycolipid andprotein interactions [59] by incorporating natural and synthet-ic glycolipids into liposomes, which were immobilized ontoan azide functionalized glass slide for microarray application(Fig. 6b). Briefly, liposomes with DDPE-PEG2000-TP andgangliosides were prepared and printed on azide functional-ized glass slide to study the interactions between glycolipidsand lectins. These microarrays mimic the presentation of car-bohydrates on the cell surface. The lipid bilayer microenvi-ronment impacts the modulation of the glycolipid presentationand the fluidity of the lipid bilayer allows optimal geometricpositioning of the glycan head groups. In our study, the intactliposomes with natural glycolipids GM1 and GM3 wereprinted on an azide functionalized glass slide; liposomalGM1 was immobilized at different range of concentrationsfrom 10 to 400 μg/mL and incubated with FITC-cholera toxinB (CTB). Also, the GM1 liposomal microarray was incubated

with different concentrations of CTB. The binding of CTB toGM1 liposome increased with increase in the concentration ofGM1 as well as CTB but reached saturation at 60 μg/mL ofCTB indicating that the binding is density dependent. Also,the GM1 microarray was incubated with lectins MAA andPNA, which binds specifically to α(2-3)-sialic acid and β-galactose, respectively. The binding curve reached saturationat 100 μg/mL for both MAA and PNA. GM3 microarray wasalso incubated with MAA and observed that the bindingreached saturation at 100 μg/mL. These binding curves indi-cate that the liposomal microarrays with different glycolipidsshow specific density dependent binding capacity and affinityas well. The liposomal glyco-microarrays have advantages ofproviding mobility to glycans by closely mimicking the cellsurface glycans and also controlling the surface glycan densityin an immobilized state and thus provide an important tool tostudy cell surface carbohydrate functions.

a

b

Fig. 6 Glyco-liposome-based glycoarray: a Surface functionalization, immobilization and glyco-functionalization of liposome via Staudinger ligation[30], b liposomal glyco-microarray formation based immobilization via Staudinger ligation [59]. Modified from Ref. [30, 59]

492 Glycoconj J (2015) 32:483–495

Glyco-nanoparticle-based glycan array

Nanoparticles have been used enormously for biomedical ap-plications as the size of the nanoparticles is close to that ofbiomolecules [60]. Glyco-nanoparticles (GNPs), sugar-coatednanoparticles, were synthesized in the past decade to study theinteractions of carbohydrate and carbohydrate binding mole-cules [61]. Three major types of nanoparticles functionalizedwith glycans have been reported so far: gold and silverglyconanoparticles, semiconductor glyco-quantum dots andmagnetic glyconanoparticles.

Yan et al. recently reported glyco-nanoparticle (GNP)-based microarrays [34], where glyco-nanoparticles wereprinted on a photoactive surface followed by covalentimmobilization by light activation. Briefly, GNPs weresynthesized and functionalized with perfluoropernylazide(PFPA) followed by conjugating five different carbohy-drates to GNPs using photocoupling chemistry. Epoxyfunctionalized wafers were treated with PAAm-PFPA,then the GNPs were printed followed by coating the sur-face with polystyrene (PS) or poly(2-ethyl-2-oxazoline)(PEOX) to avoid nonspecific binding of lectins to thesurface (Fig. 7). These glycan microarray surfaces weretested with lectin Con A labeled with fluorescein-dopedsilica nanoparticle (FSNP-Con A) or FITC labeled lectins(FITC-Con A) and it was observed that the FSNP-Con Ashowed enhanced signal then the FITC-Con A. GNP mi-croarrays were compared with conventional glycan arraysby treating them with FSNP-Con A and observed highersignals for high affinity ligands Man3, Man2 on GNPmicroarray than in carbohydrate microarray. These results

indicated that the GNP microarrays increase the affinityand selectivity when compared to carbohydrates alone onthe array.

Conclusion

Overall, glycan microarrays have been widely employed tostudy the interactions between carbohydrates and carbohy-drate binding molecules and show great potentials for bio-medical research and applications. The glycan presentationon the array surface extremely affects the binding affinityand specificity of these glycan binding molecules. Severalglyco-macroligands were developed to control the glycanpresentation on the surface in terms of multivalency, den-sity and orientation of glycan on the array surface. Theseglyco-macroligands microarray contributed greatly to un-derstand the importance of presentation of glycans on thesurface in terms of specificity and selectivity of glycanbinding molecules. However, it is still difficult to controlthe presentation of glycans precisely for every glycan bind-ing molecule on the surface as the requirement of the gly-can density and orientation changes according to the typeof glycan binding molecule of interest. The site-specificimmobilization of glyco-macrolig and is still a challengefor controlling its orientation and density on the array sur-face as well. Therefore, there is a lot of scope for improve-ment in developing methods for varying glycan presenta-tion and extending the applications of carbohydratemicroarrays.

Fig. 7 Glyconanoparticle microarray with Man 3, Man 2, Man, Glc, Fru and interactions with FSNP-Con A or FITC-Con A [34]

Glycoconj J (2015) 32:483–495 493

Acknowledgments This work was supported by American HeartAssociation Grant-in Aid (14GRANT20290002) and Faculty ResearchDevelopment Fund and the research fund from the Center for Gene Reg-ulation in Health and Disease (GRHD) at Cleveland State Universitysupported by Ohio Department of Development (ODOD). This workwas partially supported by grants from The National Natural ScienceFoundation of China (31328006). H, Nie appreciates the China OverseaScholar Award from China Scholarship Council.

References

1. Lis, H., Sharon, N.: Lectins: carbohydrate-specific proteins thatmediate cellular recognition. Chem. Rev. 98, 637–674 (1998)

2. Flitsch, S.L., Ulijn, R.V.: Sugars tied to the spot. Nature 421(6920),219–220 (2003)

3. Sodetz, J.M., Paulson, J.C., McKee, P.A.: Carbohydrate composi-tion and identification of blood group A, B, and H oligosaccharidestructures on human Factor VIII/von Willebrand factor. J. Biol.Chem. 254, 10754–10760 (1979)

4. Karlsson, K.-A.: Microbial recognition of target-cellglycoconjugates. Curr. Opin. Struct. Biol. 5(5), 622–635 (1995)

5. Lee, K.J., Mao, S., Sun, C., Gao, C., Blixt, O., Arrues, S., Hom,L.G., Kaufmann, G.F., Hoffman, T.Z., Coyle, A.R., Paulson, J.,Felding-Habermann, B., Janda, K.D.: Phage-display selection of ahuman single-chain Fv antibody highly specific for melanoma andbreast cancer cells using a chemoenzymatically cynthesized GM3-carbohydrate Antigen. J. Am. Chem. Soc. 124, 12439–12446(2002)

6. Varki, A.: Biological roles of oligosaccharides: all of the theoriesare correct. Glycobiology 3(2), 97–130 (1993)

7. Park, S., Shin, I.: Fabrication of carbohydrate chips for studyingprotein-carbohydrate interactions. Angew. Chem. 114(17), 3312–3314 (2002)

8. Fukui, S., Feizi, T., Galustian, C., Lawson, A.M., Chai, W.:Oligosaccharide microarrays for high-throughput detection andspecificity assignments of carbohydrate-protein interactions. Nat.Biotechnol. 20(10), 1011–1017 (2002)

9. Schwarz, M., Spector, L., Gargir, A., Shtevi, A., Gortler, M.,Altstock, R.T., Dukler, A.A., Dotan, N.: A new kind of carbohy-drate array, its use for profiling antiglycan antibodies, and the dis-covery of a novel human cellulose-binding antibody. Glycobiology13(11), 749–754 (2003)

10. Disney,M.D., Seeberger, P.H.: The use of carbohydratemicroarraysto study carbohydrate-cell interactions and to detect pathogens.Chem. Biol. 11(12), 1701–1707 (2004)

11. Oyelaran, O., Li, Q., Farnsworth, D., Gildersleeve, J.C.:Microarrays with varying carbohydrate density reveal distinct sub-populations of serum antibodies. J. Proteome Res. 8(7), 3529–3538(2009)

12. Qiu, Y., Patwa, T.H., Xu, L., Shedden, K., Misek, D.E., Tuck, M.,Jin, G., Ruffin, M.T., Turgeon, D.K., Synal, S., Bresalier, R.,Marcon, N., Brenner, D.E., Lubman, D.M.: Plasma glycoproteinprofiling for colorectal cancer biomarker identification by lectinglycoarray and lectin blot. J. Proteome Res. 7(4), 1693–1703(2008)

13. Zhang, Y., Li, Q., Rodriguez, L.G., Gildersleeve, J.C.: An array-based method to identify multivalent inhibitors. J. Am. Chem. Soc.132(28), 9653–9662 (2010)

14. Park, S., Gildersleeve, J.C., Blixt, O., Shin, I.: Carbohydrate micro-arrays. Chem. Soc. Rev. 42(10), 4310–4326 (2013)

15. Wang, L., Cummings, R.D., Smith, D.F., Huflejt, M., Campbell,C.T., Gildersleeve, J.C., Gerlach, J.Q., Kilcoyne, M., Joshi, L.,Serna, S., Reichardt, N.C., Parera, P.N., Pieters, R.J., Eng, W.,

Mahal, L.K.: Cross-platform comparison of glycan microarray for-mats. Glycobiology 24(6), 507–517 (2014)

16. Turnbull, W.B., Stoddart, J.F.: Design and synthesis ofglycodendrimers. Rev. Mol. Biotechnol. 90(3–4), 231–255 (2002)

17. Roy, R.: Glycodendrimers: a new class of biopolymers. Polym.News 21(7), 226–232 (1996)

18. Jayaraman, N., Nepogodiev, S.A., Stoddart, J.F.: Syntheticcarbohydrate-containing dendrimers. Chem. Eur. J. 3(8), 1193–1199 (1997)

19. Branderhorst, H.M., Ruijtenbeek, R., Liskamp, R.M.J., Pieters,R.J.: Multivalent carbohydrate recognition on a glycodendrimer-functionalized flow-through chip. Chembiochem 9(11), 1836–1844 (2008)

20. Parera Pera, N., Branderhorst, H.M., Kooij, R., Maierhofer, C., vander Kaaden, M., Liskamp, R.M.J., Wittmann, V., Ruijtenbeek, R.,Pieters, R.J.: Rapid screening of lectins for multivalency effectswith a glycodendrimer microarray. Chembiochem 11(13), 1896–1904 (2010)

21. Zhou, X., Turchi, C., Wang, D.: Carbohydrate cluster microarraysfabricated on three-dimensional dendrimeric platforms for function-al glycomics exploration. J. Proteome Res. 8(11), 5031–5040(2009)

22. Fukuda, T., Onogi, S., Miura, Y.: Dendritic sugar-microarrays byclick chemistry. Thin Solid Films 518(2), 880–888 (2009)

23. Spain, S.G., Gibson, M.I., Cameron, N.R.: Recent advances in thesynthesis of well-defined glycopolymers. J Polym. Sci. A Polym.Chem. 45, 2059–2072 (2007)

24. Godula, K., Rabuka, D., Nam, K.T., Bertozzi, C.R.: Synthesis andmicrocontact printing of dual end-functionalized mucin-likeglycopolymers for microarray applications. Angew. Chem. Int.Ed. 48(27), 4973–4976 (2009)

25. Godula, K., Bertozzi, C.R.: Synthesis of glycopolymers for micro-array applications via ligation of reducing sugars to a poly(acryloylhydrazide) scaffold. J. Am. Chem. Soc. 132(29), 9963–9965 (2010)

26. Lee, S.-G., Brown, J.M., Rogers, C.J., Matson, J.B.,Krishnamurthy, C., Rawat, M., Hsieh-Wilson, L.C.: End-functionalized glycopolymers as mimetics of chondroitin sulfateproteoglycans. Chem. Sci. 1(3), 322–325 (2010)

27. Narla, S.N., Sun, X.-L.: Orientated glyco-macroligand formationbased on site-specific immobilization of O-cyanate chain-end func-tionalized glycopolymer. Org. Biomol. Chem. 9, 845–850 (2011)

28. Narla, S.N., Sun, X.-L.: Glyco-macroligand microarray with con-trolled orientation and glycan density. Lab Chip 12(9), 1656–1663(2012)

29. Zhu, X.Y., Holtz, B., Wang, Y., Wang, L.-X., Orndorff, P.E., Guo,A.: Quantitative glycomics from fluidic glycan microarrays. J. Am.Chem. Soc. 131, 13646–13650 (2009)

30. Ma, Y., Zhang, H., Gruzdys, V., Sun, X.-L.: Azide-reactive lipo-some for chemoselective and biocompatible liposomal surfacefunctionalization and glyco-liposomal microarray fabrication.Langmuir 27(21), 13097–13103 (2011)

31. Zhang, H., Ma, Y., Sun, X.-L.: Chemically selective liposome sur-face glyco-functionalization. Methods Mol. Biol. 751, 269–280(2011)

32. Zhang, H., Ma, Y., Sun, X.-L.: Chemically-selective surface glyco-functionalization of liposomes through Staudinger ligation. Chem.Commun. 21, 3032–3034 (2009)

33. Zhang, Y., Campbell, C., Li, Q., Gildersleeve, J.C.:Multidimensional glycan arrays for enhanced antibody profiling.Mol. BioSyst. 6(9), 1583–1591 (2010)

34. Tong, Q., Wang, X., Wang, H., Kubo, T., Yan, M.: Fabrication ofglyconanoparticle microarrays. Anal. Chem. 84(7), 3049–3052(2012)

35. Matthews, O.A., Shipway, A.N., Stoddart, J.F.: Dendrimers-branching out from curiosities into new technologies. Prog.Polym. Sci. 23, 1–56 (1998)

494 Glycoconj J (2015) 32:483–495

36. Yalong, Z., Gildersleeve, J.C.: General procedure for the synthesisof neoglycoproteins and immobilization on epoxide-modified glassslides. Methods Mol. Biol. 808, 155–165 (2012)

37. Kerékgyártó, M., Fekete, A., Szurmai, Z., Kerékgyártó, J., Takács,L., Kurucz, I., Guttman, A.: Neoglycoproteins as carbohydrate an-tigens: synthesis, analysis, and polyclonal antibody response.Electrophorosis 34, 2379–2386 (2013)

38. Narla, S.N., Nie, H., Li, Y., Sun, X.-L.: Recent advances in thesynthesis and biomedical applications of chain-end functionalizedglycopolymers. J. Carbohyr. Chem. 31(2), 67–92 (2012)

39. Godula, K., Bertozzi, C.R.: Density variant glycan microarray forevaluating cross-linking of mucin-like glycoconjugates by lectins.J. Am. Chem. Soc. 134, 15732–15742 (2012)

40. Bian, S., Zieba, S.B., Morris, W., Han, X., Richter, D.C., Brown,K.A., Mirkin, C.A., Braunschweig, A.B...: Beam pen lithographyas a new tool for spatially controlled photochemistry, and its utili-zation in thesynthesis of multivalent glycan arrays. Chem. Sci. 5,2023–2030 (2014)

41. Chan, Y.-H.M., Boxer, S.G.: Model membrane systems and theirapplications. Curr. Opin. Chem. Biol. 11(6), 581–587 (2007)

42. Denning, E.J., Beckstein, O.: Influence of lipids on protein-mediated transmembrane transport. Chem. Phys. Lipids 169, 57–71 (2013)

43. Chrai, S., Murari, R., Imran, A.: Liposomes: a review. BioPharm14(11), 10–14 (2001)

44. Akbarzadeh, A., Rezaei-Sadabady, R., Davaran, S., Joo, S.W.,Zarghami, N., Hanifehpour, Y., Samiei, M., Kouhi, M., Nejati-Koshki, K.: Liposome: classification, preparation, and applications.Nanoscale Res. Lett. 8, 102–110 (2013)

45. Patil, Y.P., Jadhav, S.: Novel methods for liposome preparation.Chem. Phys. Lipids 177, 8–18 (2014)

46. Tamm, L.K., McConnell, H.M.: Supported phospholipid bilayers.Biophys. J. 47(1), 105–113 (1985)

47. Granéli, A., Rydström, J., Kasemo, B., Höök, F.: Formation of sup-ported lipid bilayer membranes on SiO2 from proteoliposomes con-taining transmembrane proteins. Langmuir 19(3), 842–850 (2003)

48. Cremer, P.S., Boxer, S.G.: Formation and spreading of lipid bilayerson planar glass supports. J. Phys. Chem. B 103(13), 2554–2559(1999)

49. Richter, R.P., Brisson, A.R.: Following the formation of supportedlipid bilayers on mica: a study vombining AFM, QCM-D, andellipsometry. Biophys. J. 88(5), 3422–3433 (2005)

50. Richter, R.P., Bérat, R., Brisson, A.R.: Formation of solid-supported lipid bilayers: an integrated view. Langmuir 22(8),3497–3505 (2006)

51. Tabaei, S.R., Jönsson, P., Brändén, M., Hook, F.: Self-assemblyformation of multiple DNA-tethered lipid bilayers. J. Struct. Biol.168(1), 200–206 (2009)

52. Silin, V.I.,Wieder, H.,Woodward, J.T., Valincius, G., Offenhausser,A., Plant, A.L.: The role of surface free energy on the formation ofhybrid bilayer membranes. J. Am. Chem. Soc. 124, 14676–14683(2002)

53. Terrettaz, S., Mayer, M., Vogel, H.: Highly electrically insulatingtethered lipid bilayers for probing the function of ion channel pro-teins. Langmuir 19, 5567–5569 (2003)

54. Munro, J.C., Frank, C.W.: In situ formation and characterization ofpoly(ethylene glycol)-supported lipid bilayers on gold surfaces.Langmuir 20(24), 10567–10575 (2004)

55. Wagner, M.L., Tamm, L.K.: Reconstituted syntaxin1A/SNAP25interacts with negatively charged lipids as measured by lateral dif-fusion in planar supported bilayers. Biophys. J. 81(1), 266–275(2001)

56. Goennenwein, S., Tanaka, M., Hu, B., Moroder, L., Sackmann, E.:Functional incorporation of integrins into solid supported mem-branes on ultrathin films of cellulose: impact on adhesion.Biophys. J. 85, 646–655 (2003)

57. Yoshina-Ishii, C., Boxer, S.G.: Arrays of mobile tethered vesicleson supported lipid bilayers. J. Am. Chem. Soc. 125(13), 3696–3697(2003)

58. Svedhem, S., Pfeiffer, I., Larsson, C., Wingren, C., Borrebaeck, C.,Höök, F.: Patterns of DNA-labeled and scFv-antibody-carrying lip-id vesicles directed by material-specific immobilization of DNAand supported lipid bilayer formation on an Au/SiO2 template.Chembiochem 4(4), 339–343 (2003)

59. Ma, Y., Sobkiv, I., Gruzdys, V., Zhang, H., Sun, X.-L.: Liposomalglyco-microarray for studying glycolipid–protein interactions.Anal. Bioanal. Chem. 404, 51–58 (2012)

60. Murthy, S.K.: Nanoparticles in modern medicine: state of theart and future challenges. Int. J. Nanomed. 2(2), 129–141(2007)

61. Marradi, M., Chiodo, F., García, I., Penadés, S.: Glyconanoparticlesas multifunctional and multimodal carbohydrate systems. Chem.Soc. Rev. 42(11), 4728–4745 (2013)

Glycoconj J (2015) 32:483–495 495